Vehicle type identification method and system

Abstract

The invention discloses a vehicle type identification method and system. The method comprises a process of classifier generation by machine training and a process of determination of a to-be-measured picture. During the process of classifier generation, an image range needed in a picture of a training set is determined based on a licence plate; the determined image range is divided into several areas; feature information in all areas is selected; all feature information in respective selected areas are inputted into a machine training unit to generate classifiers corresponding to all areas one by one; single-area determination is carried out on a to-be-measured picture by the generated classifiers; and according to the single-area determination result, a vehicle type identification result is obtained by multi-area confidence coefficient fusion determination. With the method and system, the precision of vehicle type identification is enhanced effectively and detailed information like a vehicle manufacture can be identified. The method and system can be applied to the intelligent traffic system especially and the strong evidence can be provided for dealing with a traffic accident.

Description

technical field [0001] The invention relates to the technical field of vehicle identification in video image processing. Background technique [0002] Vehicle type recognition is an important part of the intelligent transportation system, which can provide strong evidence for the handling of traffic incidents, and can provide more help for functions such as vehicle tracking. There are mainly two types of car model recognition in the prior art, namely template-based recognition and classifier-based recognition. [0003] Among them, template-based recognition is suitable for relatively specific scenarios. A Chinese patent with application number 201410014474.1 discloses a method and device for vehicle vehicle identification in ETC lanes, which is suitable for non-parking electronic toll collection systems. Vehicle models in this scenario The recognition only requires that the recognized vehicle model can correspond to the template stored in the system, and the distance betwee...

AI Technical Summary

Problems solved by technology

[0003] Among them, template-based recognition is suitable for relatively specific scenarios. A Chinese patent with application number 201410014474.1 discloses a method and device for vehicle vehicle identification in ETC lanes, which is suitable for non-parking electronic toll collection systems. Vehicle models in this scenario Recognition only requires that the recognized vehicle model can correspond to the template stored in the system, and the distance between the recognized vehicle's environment and the camera is relatively fixed, but this method cannot handle vehicle model recognition in complex environments

[0004] For this reason, another Chinese patent with the application number of 201210049730.1 discloses a car model recognition method in complex scenes. The principle adopted is the classifier-based recognition method mentioned above, which mainly involves selecting features and combining specific The machine learning discriminant method is used for classification, and the selection of features and the classifier generated by machine learning are the key to determining the efficiency and accuracy of vehicle recognition. In this scheme, the selection of features and the generation of classifiers can meet the requirements for small cars and The identification between large vehicles can also satisfy the identification between cars and jeeps, taxis and vans, etc., but specific identification such as manufacturer and year cannot be achieved

Images